If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Cardio Metabolic Unit, Department of Medicine and Department of Laboratory Medicine, Karolinska Institutet, and Medicine Unit Endocrinology, Theme Inflammation and Ageing, Karolinska University Hospital, Stockholm, Sweden
Mutations in the LCAT gene cause familial LCAT deficiency (Online Mendelian Inheritance in Man ID: #245900), a very rare metabolic disorder. LCAT is the only enzyme able to esterify cholesterol in plasma, whereas sterol O-acyltransferases 1 and 2 are the enzymes esterifying cellular cholesterol in cells. Despite the complete lack of LCAT activity, patients with familial LCAT deficiency exhibit circulating cholesteryl esters (CEs) in apoB-containing lipoproteins. To analyze the origin of these CEs, we investigated 24 carriers of LCAT deficiency in this observational study. We found that CE plasma levels were significantly reduced and highly variable among carriers of two mutant LCAT alleles (22.5 [4.0–37.8] mg/dl) and slightly reduced in heterozygotes (218 [153–234] mg/dl). FA distribution in CE (CEFA) was evaluated in whole plasma and VLDL in a subgroup of the enrolled subjects. We found enrichment of C16:0, C18:0, and C18:1 species and a depletion in C18:2 and C20:4 species in the plasma of carriers of two mutant LCAT alleles. No changes were observed in heterozygotes. Furthermore, plasma triglyceride-FA distribution was remarkably similar between carriers of LCAT deficiency and controls. CEFA distribution in VLDL essentially recapitulated that of plasma, being mainly enriched in C16:0 and C18:1, while depleted in C18:2 and C20:4. Finally, after fat loading, chylomicrons of carriers of two mutant LCAT alleles showed CEs containing mainly saturated FAs. This study of CEFA composition in a large cohort of carriers of LCAT deficiency shows that in the absence of LCAT-derived CEs, CEs present in apoB-containing lipoproteins are derived from hepatic and intestinal sterol O-acyltransferase 2.
Cholesteryl esters (CEs) are synthesized in humans by three distinct enzymes, LCAT, sterol O-1 acyltransferase (SOAT1), and sterol O-2 acyltransferase (SOAT2)—previously denominated ACAT1 and ACAT2. LCAT acts in plasma and other biological fluids, where it catalyzes the transacylation of the sn-2 FA of lecithin, which usually is polyunsaturated, to the free 3-OH group of cholesterol, generating CE and lysolecithin (
). Here, it participates to the regulation of intracellular cholesterol homeostasis and to the synthesis of CEs fated to apoB-containing lipoprotein secretion (VLDL and chylomicrons). The most commonly SOAT2-derived CEs are cholesteryl oleate (C18:1) and cholesteryl palmitate (C16:0) (
). Immediately after entering the circulation, VLDL will become substrate of LCAT, which will produce the majority of circulating CEs. Despite the unchanged or increased levels of apoB in circulation, Soat2 deficiency in mice prevents atherosclerosis because of decreased content in VLDL-CEs and LDL-CEs and to a shift of lipoprotein cholesteryl ester FA (CEFA) composition toward the unsaturated forms (
Liver-specific inhibition of acyl-coenzyme a:cholesterol acyltransferase 2 with antisense oligonucleotides limits atherosclerosis development in apolipoprotein B100-only low-density lipoprotein receptor-/- mice.
Mutations in the gene coding for the LCAT enzyme cause the very rare recessive metabolic disorder called familial LCAT deficiency (FLD, Online Mendelian Inheritance in Man ID: #245900). Patients with FLD have severe HDL deficiency, hypertriglyceridemia, and an increased unesterified to total cholesterol ratio (
), we have shown that despite the absence of LCAT activity, carriers have circulating CEs, highly variable among subjects and of unknown origin. In the present study, we have evaluated the origin of circulating CEs in carriers of LCAT deficiency, testing the hypothesis that SOAT2 can contribute to the level of plasma CEs in humans.
Materials and methods
Twenty-four carriers of LCAT gene mutations, including eight carriers of two mutant LCAT alleles and seven heterozygotes, and nine noncarrier family members, all belonging to the Italian LCAT-deficient families (
), were included in this observational study. Selection of subjects was conditioned by plasma sample availability. A detailed overview of the LCAT mutations can be found in supplemental Table S1. All procedures were followed in accordance with the ethical standards of the local institutional committees on human experimentation and according to the Helsinki Declaration of 1964, as revised in 2013. The study was approved by the institutional Ethic Committees (approval number: 446-092014). All subjects were fully informed about the modalities of the study and enrolled after signed informed consent.
Plasma biochemical analyses
Blood samples were collected after an overnight fast in Na2-EDTA tubes, plasma was separated by low-speed centrifugation at 4°C within a few minutes from collection, and then immediately frozen. Plasma total and HDL cholesterol, unesterified cholesterol (UC), triglyceride (TG), phospholipid (PL), and apolipoprotein levels were determined with certified methods using a Roche c311 autoanalyzer (Roche Diagnostics) (
). The amount of CEs was calculated by subtracting UC from total cholesterol, and the difference was multiplied by 1.68 to have a precise estimation of the CE mass. LDL and non-HDL cholesterol were calculated by the Friedewald’s equation and as the difference between total cholesterol and HDL-cholesterol, respectively. In the case of TG levels >400 mg/dl, LDL cholesterol was assessed using a direct measurement method. Plasma CE and PL FA composition (CEFA, PLFA) was measured by a gas-liquid chromatographic method (
). TG species were detected as neutral loss of one FA chain. FA distribution was determined from molecular mass, FA chain loss, and retention time. Hepatic and intestinal SOAT2 contribution to CE formation was evaluated by analyzing CEFA composition in VLDL and chylomicrons, respectively, as described below.
Lipoprotein purification and analysis
Plasma lipoproteins for the composition analysis were separated by sequential ultracentrifugation using a Beckman TL 100 ultracentrifuge equipped with a TL 100.3 rotor (Beckman Instruments, Fullerton, CA). Lipoproteins were dialyzed against saline to remove potassium bromide, and lipoprotein fractions were shown to be free of other lipoproteins by electrophoresis in agarose gel. Total cholesterol, UC, PL, and TG contents of lipoprotein fractions were determined by enzyme techniques; the CE mass was calculated as aforementioned. Protein content was measured by the method of Lowry using bovine serum albumin as standard.
VLDLs for the FA analysis were separated by fast-protein liquid chromatography using a single Superose 6 HR10/30 column (Pharmacia). Plasma samples (500 μl) were applied to the column and eluted in Tris buffer, and fractions were collected and pooled according to the elution times for VLDL, LDL, and HDL. Lipoprotein fractions were shown to be free of other lipoproteins by electrophoresis in agarose gel. Then, CEFA composition of isolated VLDL was determined as described aforementioned.
Two homozygous FLD patients and two matched controls received, after an overnight fast, a test meal consisting of 2 eggs, 40 g butter, 40 g olive oil, 2 slices of bacon, and 2 salted brioches (1,478 kcal, fat 73% of total energy). The test meal was consumed during a period of 15 min, and the subjects were deprived of any source of energy for the next 6 h. Blood samples were collected in Na2-EDTA (1 mg/ml) tubes before the test meal (0 h) and after 1, 2, 3, 4, and 6 h, and plasma was separated by low-speed centrifugation at 4°C. Plasma lipid/apolipoproteins were measured as described above. Chylomicrons were immediately separated from fresh plasma by ultracentrifugation at 50,000 rpm for 20 min at 4°C, using a Beckman TL 100 ultracentrifuge equipped with a TL 100.3 rotor (
). Plasma lipoproteins were separated by fast-protein liquid chromatography (see above). CEFA composition was determined in chylomicrons at 1, 2, 3, 4, and 6 h after meal as described above. Since CEFA distribution in plasma was not different to that of controls, high-fat load test was not performed in carriers of one mutant LCAT allele. Test meal nutrient composition is reported in supplemental Table S2.
Data were summarized as median (first and third quartiles) and with n (%) for categorical variables, if not otherwise stated. The data were analyzed according to what were reported in figure and table legends. The number of samples for each test is reported in the figure or table.
The association of biochemical parameters with LCAT genotype was assessed by Spearman’s rho correlation versus the number of mutant LCAT alleles (0, 1, or 2). Comparisons between groups were performed by Kruskal-Wallis nonparametric test followed by Dunn’s correction for multiple comparisons or Mann-Whitney test, where appropriate.
GraphPad Prism 9 (GraphPad Software, Inc) was used for graphic presentation. All statistical analyses were performed using SPSS software, version 27.0 (SPSS, Inc, Chicago, IL). Tests were two-sided, and P values <0.05 were considered as statistically significant.
Lipoprotein composition is significantly altered in carriers of LCAT mutations
The characteristics of the subjects participating in the study are summarized in Table 1. Carriers and controls were comparable for age and BMI. As previously reported (
), carriers of LCAT deficiency had lower LDL cholesterol and HDL cholesterol levels and higher TG than control subjects, whereas non-HDL cholesterol was similar in carriers and controls. Apolipoprotein B was significantly lower in carriers of two mutant LCAT alleles, making the non-HDL cholesterol/apolipoprotein B ratio much higher in carriers of two mutant LCAT alleles compared with heterozygous carriers and controls (2.38 [1.84–3.66], 1.41 [1.26–1.49], and 1.48 [1.40–1.55], respectively, P = 0.081). The TG/non-HDL cholesterol ratio was also significantly higher in carriers of two mutant LCAT alleles compared with heterozygous carriers and controls (2.33 [1.33–3.30], 1.11 [0.89–1.26], and 0.78 [0.62–1.08], respectively, P < 0.0001).
Table 1Demographic and lipid/lipoprotein data in carriers of LCAT mutations and controls
Carriers of two mutant LCAT alleles
Carriers of one mutant LCAT allele
Total cholesterol (mg/dl)
Non-HDL cholesterol (mg/dl)
LDL cholesterol (mg/dl)
HDL cholesterol (mg/dl)
Apolipoprotein A-I (mg/dl)
Apolipoprotein A-II (mg/dl)
Apolipoprotein B (mg/dl)
LCAT activity (nmol/ml/h)
LCAT mass (μg/ml)
Data are reported as median (first and third quartiles) or numbers.
P values reported are calculated by Spearman’s rho correlation.
Consistently with the presence of pathogenic LCAT mutations, carriers showed significantly reduced LCAT mass and activity with a gene dose-dependent effect. Notably, LCAT activity was undetectable in all carriers of two mutant LCAT alleles, whereas in heterozygous carriers, it was half of controls. As a consequence of complete LCAT deficiency, cholesterol esterification was compromised in carriers (Table 1), and this was particularly evident in carriers of two mutant LCAT alleles. Indeed, they showed markedly increased plasma UC levels and unesterified/total cholesterol ratio. Surprisingly, CEs were detectable in plasma, although very low (88% lower compared with controls). As described for the entire Italian cohort (
), CEs in carriers of LCAT mutations display high variability, ranging from 4 to 87 mg/dl, and were significantly correlated with total cholesterol levels (r = 0.756; P = 0.030), independently of the causative mutation (supplemental Fig. S1). Consistently, CEs to total cholesterol ratio was significantly lower in carriers of two mutant LCAT alleles compared with controls (Table 1), without appreciable differences between females and males (0.13 [0.06] vs. 0.17 [0.06–0.23], respectively, P = 1.000). Carriers of one mutant LCAT allele had normal unesterified/total cholesterol ratio and plasma UC levels, whereas plasma CEs were slightly reduced compared with controls (Table 1).
The percentage of lipoprotein composition was measured in a representative subgroup of carriers and controls and reported in Figure 1. Lipoprotein composition was significantly affected by the presence of LCAT gene mutations (Fig. 1): both VLDL and LDL from carriers of two mutant alleles were characterized by an increased content of UC compared with control lipoproteins (9.3 ± 3.7% vs. 4.2 ± 1.4%, P < 0.001 for VLDL; 18.8 ± 3.9% vs. 8.9 ± 2.0%, P < 0.001 for LDL). Parallelly, VLDL and LDL showed a significantly diminished content of CEs compared with controls (2.3 ± 2.1% vs. 8.4 ± 3.6%, P = 0.010 for VLDL; 5.2 ± 4.4 vs. 37.3 ± 5.1%, P < 0.0001 for LDL). CEs were almost undetectable in HDL from carriers of two mutant LCAT alleles (0.3 ± 0.7% vs. 21.5 ± 3.6%, P < 0.0001 vs. controls), suggesting that the little amount of circulating CEs in complete LCAT deficiency is carried only by apoB-containing lipoproteins. LDL from carriers of two mutant LCAT alleles displayed a very high content of PL compared with controls (P < 0.001, Fig. 1), confirming the presence of abnormal lipoproteins, that is, lipoprotein X (
), large vesicles that form to accommodate the excess of circulating UC and has a density similar to LDL. TG content was also increased in LDL from carriers of two mutant LCAT alleles (22.7 ± 6.1% vs. 8.6 ± 2.6% in control LDL, P < 0.0001).
The composition of VLDL from carriers of one mutant LCAT allele was almost identical to that of controls. Both carriers’ LDL and HDL showed instead a higher content in TG than controls (14.9 ± 3.2% vs. 8.6 ± 2.6, P = 0.037 for LDL and 5.6 ± 1.1 vs. 3.4 ± 1.1, P = 0.011 for HDL).
CEFA composition in whole plasma and VLDL particles indicates the SOAT2 origin of circulating CEs
The percentage of FA composition of plasma CEs was measured, and the data are shown in supplemental Table S3 and Fig. 2. Plasma CEFA distribution showed an enrichment in saturated (C16:0 and C18:0) and monounsaturated (C18:1) FAs and a depletion in polyunsaturated FAs (C18:2, C20:4) in carriers of two mutant LCAT alleles. In addition, a significant increased plasma oleate/linoleate ratio (C18:1/C18:2) was observed compared with controls (4.71 [1.31–6.02] vs. 0.55 [0.51–0.60], P < 0.002). No significant changes were instead detected in carriers of one mutant LCAT allele. CEFA composition clearly indicates that they are SOAT derived. Both SOAT1 and SOAT2 use unsaturated and monounsaturated as preferential FA for CE formation, but only SOAT2 participates in the formation of CEs that are incorporated in apoB-containing lipoproteins (
). Thus, circulating CEs in LCAT deficiency are SOAT-2 derived.
To exclude a possible role of the diet, TGFA composition was also assessed and reported in supplemental Table S4. Plasma TGFA distribution was remarkably similar between carriers of LCAT deficiency and controls (supplemental Table S4). PL FAs were also analyzed and showed only a reduced C20:4 content in carriers of two mutant LCAT alleles compared with controls (supplemental Table S5).
CEFA profile in VLDL essentially recapitulated that of plasma (Table 2). VLDLs from carriers of two mutant LCAT alleles were mainly enriched in C16:0 and C18:1, whereas were depleted in C18:2 and C20:4. The reduced content of C18:2 altered the CE oleate/linoleate ratio in VLDL: an increase of C18:1/C18:2 ratio was observed in subjects carrying two mutant LCAT alleles, consistent with what observed on whole plasma (Table 2). As observed in plasma, CEFA distribution in VLDL from carriers of one mutant LCAT allele was comparable to controls (Table 2).
Table 2VLDL-CEFA distribution in carriers of LCAT gene mutations and controls
CEFA composition in chylomicrons supports the SOAT2 origin of postprandial CE
To dissect the contribution of intestinal SOAT2 to CEs in circulation in the postprandial state, we analyzed FA composition of chylomicron CE (chylomicron CEFA) after a fat test meal. Figure 3 reported chylomicron-CEFA composition at TG peak, which corresponded to 2 h after meal for control subjects and 3 h after meal for carriers of two mutant LCAT alleles. In patients with LCAT deficiency, the percentage distribution of chylomicron CEs containing saturated, monounsaturated, or polyunsaturated FA was shifted toward the saturated ones, when compared with control chylomicrons (Fig. 3). The enrichment in palmitate (C16:0), stearate (C18:0), and oleate (C18:1) and the depletion in linoleate (C18:2) and arachidonate (C20:4) was consistent with what observed on whole plasma and in VLDL (supplemental Table S3 and Table 2). Chylomicron-CEFA distribution was almost unchanged with time after eating, both in carriers and controls (supplemental Fig. S2). Of note, CEFA distribution in FLD and controls were very similar in fasting and postprandial state, suggesting a comparable biochemical fingerprint of SOAT2 in liver and intestine (Figs. 2 and 3).
The present study demonstrates that i) patients with familial LCAT deficiency have circulating plasma CEs, despite the complete absence of LCAT activity, which are only detected in apoB-containing lipoproteins but not in HDL and ii) the FA composition of circulating CEs indicates that they are SOAT2 derived.
Familial LCAT deficiency is a very rare recessive disorder of lipoprotein metabolism caused by loss-of-function mutations in the LCAT gene (
). LCAT is the only enzyme able to esterify cholesterol in plasma, and around 70% of the circulating cholesterol is in the esterified form. The absence of LCAT thus has a dramatic impact on circulating lipoproteins, as clearly demonstrated by the lipoprotein composition reported in the present study. In patients with FLD, circulating HDLs have virtually no CEs and in fact are discoidal particles (
), whereas VLDL and LDL contain a certain amount of CEs. The absence of CEs in carriers’ HDL indicates that CE transfer protein does not operate lipid transfer between apoB-containing lipoproteins and HDL, likely because of the absence of core lipids in HDL.
The FA composition of VLDL-CEs demonstrates that they are produced by hepatic SOAT2. Cholesteryl oleate is largely predominant followed by cholesteryl palmitate (C16:0) in VLDL from patients with FLD, whereas in control VLDL, the most abundant is cholesteryl linoleate (C18:2), the enzymatic product of LCAT. These findings together with what observed in mice, where the deletion of both Lcat and Soat2 genes leads to the complete absence of plasma CEs (
), support the concept that SOAT2 is the enzyme producing CEs fated to secretion in lipoproteins.
Plasma levels of CEs are highly variable in carriers of two mutant LCAT alleles; this variability is not attributable to the type of LCAT mutation, since LCAT activity is completely absent in all patients with FLD, but is mainly proportional to total plasma cholesterol levels, which depends on several other genetic determinants. In our cohort, we did not observe a difference in circulating CEs between males and females, despite the lower microsomal activity of SOAT2 observed in the liver of females in a small Chinese population (
). FA composition of chylomicron CEs isolated postprandially from carriers also shows to be SOAT2 derived. Interestingly, FA distribution in chylomicrons does not change postprandially in carriers, confirming the complete absence of LCAT activity.
Several studies have provided evidence that the types of CEs that predominate in plasma contribute to the relative degree of atherogenicity. In the Atherosclerosis Risk in Communities study (
), carotid intima-media thickness was positively associated with the amount of circulating SOAT2-derived CEs (cholesteryl palmitate and cholesteryl oleate), independently of other cardiovascular risk factors. A large population-based prospective study (
) provided even more robust evidence in support of the atherogenicity of SOAT2-derived CEs, showing a significant positive association between the cholesteryl oleate and cholesteryl palmitate content of plasma lipoproteins and cardiovascular disease death. Conversely, cholesteryl linoleate, the typical product of LCAT, was inversely associated with cardiovascular disease death. In line with the evidence that polyunsaturated FA containing CEs derived from LCAT are associated with decreased atherosclerosis, whereas saturated and monounsaturated CEs derived predominantly from SOAT2 are associated with increased atherosclerosis, LCAT deficiency in mice resulted in more atherosclerosis (
) have unaltered plasma levels of apoB, whereas in LCAT-deficient patients, the levels of apoB are reduced. Hence, not only the quality of CEs but also the quantity and number of particles carrying them is relevant in atherogenesis.
Arterial plaque stabilization through the use of SOAT inhibitors has been investigated for two decades but unsuccessfully (
). Whereas SOAT2 remains a potential therapeutic target for plasma cholesterol reduction and atherosclerosis prevention, targeting SOAT1 was found promising in other diseases, that is, Alzheimer’s disease, where it is implicated in the amyloid plaque development (
Some limitations of our study should be appreciated. The number of analyzed FLD patients is small and has been limited by the availability of plasma samples. Subjects have been identified among the Italian cohort of LCAT-deficient carriers, which is a large cohort for a very rare dyslipidemia. In addition, females are very few in the entire cohort and only two in the present study. Finally, the postprandial evaluation was carried out in the only two FLD patients who agreed to participate in the study.
In conclusion, this is the first unequivocal demonstration that SOAT2 contributes CEs to newly secreted VLDL and chylomicrons in humans. The LCAT-deficient model does not allow to address the potential role of SOAT2 inhibition in the treatment of atherosclerotic cardiovascular diseases, which remains to be proven.
The data that support the findings of this study are available from the corresponding author on realistic request.
The authors declare that they have no conflicts of interest with the contents of this article.
This work has been conceived, designed, and carried out with the precious collaboration of Prof Lawrence (Larry) L. Rudel, head of the Section on Lipid Sciences, Department of Pathology, Wake Forest University School of Medicine, Winston-Salem, United States. Larry Rudel passed away on August 29, 2019, and this article is dedicated to him. The authors also thank Martha Wilson, Wake Forest School of Medicine, Winston-Salem, for technical assistance. Grant from MIUR Project Excellence and the UNIMI Research Support Plan, line 2 action C. The authors are grateful to the subjects for their contribution to research.
P. P. and L. C. conceptualization; C. P., A. O., A. S., P. R., F. V., and M. L. formal analysis; C. P. data curation; C. P., P. P., and L. C. writing–original draft; C. P., A. O., A. S., P. R., F. V., M. L., P. P., and L. C. writing–review & editing.
Liver-specific inhibition of acyl-coenzyme a:cholesterol acyltransferase 2 with antisense oligonucleotides limits atherosclerosis development in apolipoprotein B100-only low-density lipoprotein receptor-/- mice.